Core Concepts
Autonomous fixed-wing aerial vehicles can safely navigate and actively map steep alpine environments, enabling efficient data collection for environmental monitoring tasks.
Abstract
This paper presents an integrated system that enables autonomous fixed-wing aerial vehicles to safely navigate and actively map steep alpine environments. The key challenges addressed include:
Safe Navigation in Steep Terrain:
The limited maneuverability of fixed-wing vehicles makes navigating close to steep terrain challenging, as they can encounter inevitable collision states.
The system uses a safe path planner that leverages circular loiter paths to plan kinematically feasible and safe trajectories.
Handling Environment Uncertainty:
Alpine environments have constantly changing conditions, including obstacles, wind, and other air traffic, which are difficult to predict.
The system integrates techniques for detecting obstacles, predicting wind fields, and avoiding other air traffic to operate safely.
Online Decision Making for Information Maximization:
The system employs an active aerial photogrammetry approach that optimizes viewpoints in real-time to maximize the quality of the reconstructed map.
This is achieved by formulating the problem as a Markov Decision Process and using Monte Carlo Tree Search to plan informative trajectories.
The system was evaluated through field tests in steep alpine terrain in Davos, Switzerland. The results demonstrate that the autonomous fixed-wing aerial vehicle can safely navigate the environment while actively mapping the region of interest, outperforming a traditional coverage-based approach.
Stats
The system has a mass of 5.7 kg and a wingspan of 2300 mm.
The target region of interest has an extent of 390 m × 295 m.
Quotes
"Monitoring large scale environments is a crucial task for managing remote alpine environments, especially for hazardous events such as avalanches."
"Fixed-wing vehicles, due to their long range and travel speeds are a promising platform to gather aerial imagery to map avalanche activities."